Study to Evaluate Safety and Efficacy of Cenegermin (Oxervate®) vs Vehicle in Severe Sjogren's Dry Eye Disease

SANDE was a short questionnaire to evaluate both dry eye intensity and frequency. It uses a 100 mm horizontal line (Visual Analogue Scale - VAS), for each of the 2 questions, to assess ocular discomfort and/or dryness. Frequency of symptoms ranged from rarely (best outcome) to all of the time (worst outcome), while the severity of symptoms ranged from very mild (best outcome) to very severe (worst outcome). Patients had to place a mark on the two given lines based on the extent of their symptoms. The locations of the marks was measured in mm from left to right and recorded as frequency and severity scores, respectively. The SANDE lines for intensity and for severity ranged from 0, being the minimal amount of dry eye symptoms (best outcome) to 100, being the maximal amount of dry eye symptoms (worst outcome). The total SANDE score was calculated by multiplying the frequency score by the severity score and obtaining the square root (0-100; the lower, the better).

Time frame: At Week 12 (Visit 5 - Follow up)

Population: Full Analysis set (FAS): The FAS population consisted of all randomized patients who received at least 1 dose of the IMP. The FAS population was used for the primary analysis of the study and to present results on efficacy data. No units other than participants were assigned.~While at baseline FAS patients were n=44 for IMP and 40 for Vehicle, at week 12 patients analyzed were n=40 in the IMP group and n=39 in the Vehicle arm.~Please note that mean is an adjusted mean.

The Schirmer test was used in ophthalmic examination to measure tear production for the diagnosis of conditions such as keratoconjunctivitis sicca and dry eye. Without previously instilling anesthetic drops, the Schirmer strip was inserted into the lower conjunctival sac at the junction of the lateral and middle thirds, avoiding touching the cornea, and the length of wetting strips in millimeters was recorded after 5 minutes. After 5 minutes had elapsed, the Schirmer's test strip was removed and the length of the tear absorption on the strip was measured (millimeters/5 minutes). Cutoff values: \<5 mm - pathologic dry eye 5-10 mm - marginal dry eye \>10 and \<30 mm - normal secretion The longer, the wetted length, the healthier the status of the eye. No units other than participants were assigned.

Time frame: At Week 4 (Visit 3)

Population: Full Analysis set The FAS population consisted of all randomized patients who received at least one dose of the investigational product. The FAS population was used for the primary analysis of the study and to present results on efficacy data. While at baseline FAS patients were n=44 for IMP and 40 for Vehicle, at week 4 patients analyzed were n=43 in the IMP group and n=39 in the Vehicle arm.

The NEI/Industry Workshop guidelines were used for grading the scale of corneal and conjunctival damage. The cornea was divided into five sectors (central, superior, inferior, nasal and temporal), each of which was scored on a scale of 0-3, with a maximal total corneal staining score of 15. Both nasally and temporally, the conjunctiva was divided into a superior paralimbal area, an inferior paralimbal area, and a peripheral area with a grading scale of 0-3 and with a maximal total score of 9 for the nasal and temporal conjunctiva (overall the total score ranged from 0-18). Briefly, grade 0 reflected normal/healthy situation, whereas grade 3 reflected a severe damage in the considered sector.

Time frame: At Week 4 (Visit 3), Week 8 (Visit 4) , Week 12 (Visit 5) and Week 16 (Visit 6)

Population: Full Analysis set (FAS): it consisted of all randomized patients who received at least one dose of the IMP. The FAS population was used for the primary analysis and to present results on efficacy data. Patients at baseline were n=44 in the cenegermin arm and n=40 in the vehicle arm. At the other timepoints after baseline, a maximum of 42 cenegermin patients and 37 vehicle patients contributed data for this outcome within the FAS population.

IDEEL was a 57-item questionnaire that assessed the impact of dry eye symptoms on everyday life. It consisted of 3 modules: * Dry eye Quality of Life (27 items) composed by 3 dimensions: 1. Impact on Daily Activities 2. Emotional Impact 3. Impact on Work Scores for each dimension of this module ranged from 0 to 100, where higher scores indicated less impact on daily activities, on work and emotions. * Treatment Satisfaction & Bother (8 items) composed by 2 dimensions: 1. Satisfaction with Treatment Effectiveness 2. Treatment-Related Bother / Inconvenience Scores for each dimension of this module range from 0 to 100, where higher scores indicate greater satisfaction with treatment effectiveness and less treatment-related bother. * Symptom Bother (20 items) composed by 1 dimension: 1. Dry Eye Symptom-Bother. Scores for the dimension of this module ranged from 0 to 100, where higher scores indicated greater symptom bother. No combination of dimensions scores was done.

Time frame: At Week 4 (Visit 3), Week 8 (Visit 4), Week 12 (Visit 5), and Week 16 (Visit 6)

Population: Full Analysis set (FAS): it consisted of all randomized patients who received at least one dose of the IMP. The FAS population was used for the primary analysis and to present results on efficacy data. Patients at baseline were n=44 in the cenegermin arm and n=40 in the vehicle arm. At the other timepoints after baseline, a maximum of 43 cenegermin patients and 39 vehicle patients contributed data for this outcome within the FAS population.

The Schirmer test was used in ophthalmic examination to measure tear production for the diagnosis of conditions such as keratoconjunctivitis sicca and dry eye. Without previously instilling anesthetic drops, the Schirmer strip was inserted into the lower conjunctival sac at the junction of the lateral and middle thirds, avoiding touching the cornea, and the length of wetting strips in millimeters was recorded after 5 minutes. After 5 minutes had elapsed, the Schirmer test strip was removed and the length of the tear absorption on the strip was measured (millimeters/5 minutes). Cutoff values: \<5 mm - pathologic dry eye 5-10 mm - marginal dry eye \>10 and \<30 mm - normal secretion The longer, the wetted length, the healthier the status of the eye.

Time frame: At Week 4 (Visit 3), Week 8 (Visit 4), Week 12 (Visit 5), and Week 16 (Visit 6)

Population: Full Analysis set (FAS): it consisted of all randomized patients who received at least one dose of the investigational product. The FAS population was used for the primary analysis of the study and to present results on efficacy data.~While at baseline FAS patients were n=44 for IMP and 40 for Vehicle, patients analyzed were:~n=43 at week 4; n=40 at weeks 8 and 12; n=41 at week 16 in the IMP arm and n=39 at weeks 4 and 12; n=40 at weeks 8 and 16 in the vehicle arm.

SANDE was a short questionnaire to evaluate both dry eye intensity and frequency. It uses a 100 mm horizontal line (Visual Analogue Scale - VAS), for each of the 2 questions, to assess ocular discomfort and/or dryness. Frequency of symptoms ranged from rarely (best outcome) to all of the time (worst outcome), while the severity of symptoms ranged from very mild (best outcome) to very severe (worst outcome). Patients had to place a mark on the two given lines based on the extent of their symptoms. The locations of the marks was measured in mm from left to right and recorded as frequency and severity scores, respectively. The SANDE lines for intensity and for severity ranged from 0, being the minimal amount of dry eye symptoms (best outcome) to 100, being the maximal amount of dry eye symptoms (worst outcome). The total SANDE score was calculated by multiplying the frequency score by the severity score and obtaining the square root (0-100; the lower, the better).

Time frame: At Week 8 (Visit 4), Week 12 (Visit 5), and Week 16 (Visit 6)

Population: Full Analysis set (FAS): it consisted of all randomized patients who received at least one dose of the IMP. The FAS population was used for the primary analysis and to present results on efficacy data. Patients at baseline were n=44 in the cenegermin arm and n=40 in the vehicle arm. At the other timepoints after baseline, a maximum of 41 cenegermin patients and 39 vehicle patients contributed data for this outcome within the FAS population.~Please note that mean is an adjusted mean.

Tear film break-up time (TFBUT) was the time taken to appear first dry spot on cornea after a complete blinking. TFBUT measurement was an easy and fast method used to assess the stability of tear film. It was a standard diagnostic procedure in the dry eye clinics. TFBUT was measured by determining the time to tear break-up. The TFBUT was performed after instillation of 5 μL of 2% preservative-free sodium fluorescein solution into the inferior conjunctival sac of each eye. The patient was instructed to blink several times to thoroughly mix the fluorescein with the tear film. A TFBUT greater than 15 seconds was considered normal, while a break time of less than 10 seconds was to be considered pathological. The shorter the time, the worse the outcome.

Time frame: At Week 4 (Visit 3), Week 8 (Visit 4), Week 12 (Visit 5), and Week 16 (Visit 6)

Population: Full Analysis set (FAS): it consisted of all randomized patients who received at least one dose of the IMP. The FAS population was used for the primary analysis and to present results on efficacy data. Patients at baseline were n=44 in the cenegermin arm and n=40 in the vehicle arm. At the other timepoints after baseline, a maximum of 43 cenegermin patients and 39 vehicle patients contributed data for this outcome within the FAS population.

IDEEL was a 57-item questionnaire that assessed the impact of dry eye symptoms on everyday life. It consisted of 3 modules: * Dry eye Quality of Life (27 items) composed by 3 dimensions: 1. Impact on Daily Activities 2. Emotional Impact 3. Impact on Work Scores for each dimension of this module ranged from 0 to 100, where higher scores indicated less impact on daily activities, on work and emotions. * Treatment Satisfaction & Bother (8 items) composed by 2 dimensions: 1. Satisfaction with Treatment Effectiveness 2. Treatment-Related Bother / Inconvenience Scores for each dimension of this module range from 0 to 100, where higher scores indicate greater satisfaction with treatment effectiveness and less treatment-related bother. * Symptom Bother (20 items) composed by 1 dimension: 1. Dry Eye Symptom-Bother Scores for the dimension of this module ranged from 0 to 100, where higher scores indicated a greater symptom bother. No combination of dimensions scores was done.

Time frame: At Week 12 (Visit 5 - Follow-up) and Week 4 (Visit 3)

Population: Full Analysis set (FAS): it consisted of all randomized patients who received at least one dose of the IMP. The FAS population was used for the primary analysis and to present results on efficacy data. Patients at baseline were n=44 in the cenegermin arm and n=40 in the vehicle arm. At the other timepoints after baseline, a maximum of 43 cenegermin patients and 39 vehicle patients contributed data for this outcome within the FAS population.

The Symptom Assessment in Dry Eye (SANDE) questionnaire was a short questionnaire to evaluate both dry eye intensity/severity and frequency. This questionnaire used a 100 mm horizontal line (Visual Analogue Scale - VAS) for each of the 2 questions to assess ocular discomfort and/or dryness experienced by the patients. In the SANDE questionnaire, frequency of symptoms ranges from rarely to all of the time and the severity of symptoms ranged from very mild to very severe. Patients were asked to place a mark on the two given lines based on the extent of their symptoms. The locations of the marks was measured in mm from left to right and recorded as frequency and severity scores, respectively. The SANDE scale ranged from 0, being the minimal amount of dry eye symptoms (best outcome) to 100, being the maximal amount of dry eye symptoms (worst outcome). In this outcome severity score at week 12 was assessed.

Time frame: At Week 12 (Visit 5 - Follow-up)

Population: Full Analysis set (FAS): The FAS population consisted of all randomized patients who received at least one dose of the investigational product. The FAS population was used for the primary analysis of the study and to present results on efficacy data.~While at baseline FAS patients were n=44 for IMP and 40 for Vehicle, at week 4 patients analyzed were n=40 in the IMP group and n=39 in the Vehicle arm.~Please note that mean is an adjusted mean.

SANDE was a short questionnaire to evaluate both dry eye intensity and frequency. It uses a 100 mm horizontal line (Visual Analogue Scale, VAS), for each of the 2 questions, to assess ocular discomfort and/or dryness. Frequency of symptoms ranged from rarely (best outcome) to all of the time (worst outcome), while the severity of symptoms ranged from very mild (best outcome) to very severe (worst outcome). Patients had to place a mark on the two given lines based on the extent of their symptoms. The locations of the marks was measured in mm from left to right and recorded as frequency and severity scores, respectively. The SANDE lines for intensity and for severity ranged from 0, being the minimal amount of dry eye symptoms (best outcome) to 100, being the maximal amount of dry eye symptoms (worst outcome). The total SANDE score was calculated by multiplying the frequency score by the severity score and obtaining the square root (0-100; the lower, the better).

Time frame: At Week 12 (Visit 5 - Follow-up)

Population: Full Analysis set (FAS): The FAS population consisted of all randomized patients who received at least one dose of the investigational product. The FAS population was used for the primary analysis of the study and to present results on efficacy data.~While at baseline FAS patients were n=44 for IMP and 40 for Vehicle, at week 4 patients analyzed were n=40 in the IMP group and n=39 in the Vehicle arm.~Please note that mean is an adjusted mean.

Tear film break-up time (TFBUT) was the time taken to appear first dry spot on cornea after a complete blinking. TFBUT measurement was an easy and fast method used to assess the stability of tear film. It was a standard diagnostic procedure in the dry eye clinics. TFBUT was measured by determining the time to tear break-up. The TFBUT was performed after instillation of 5 μL of 2% preservative-free sodium fluorescein solution into the inferior conjunctival sac of each eye. The patient was instructed to blink several times to thoroughly mix the fluorescein with the tear film. A TFBUT greater than 15 seconds was considered normal, while a break time of less than 10 seconds was to be considered pathological. Hence, the shorter the time, the worse the outcome.

Time frame: At week 4 (Visit 3) , Week 8 (Visit 4), and Week 12 (Visit 5 - Follow-up)

Population: Full Analysis set (FAS): it consisted of all randomized patients who received at least one dose of the IMP. The FAS population was used for the primary analysis and to present results on efficacy data. Patients at baseline were n=44 in the cenegermin arm and n=40 in the vehicle arm. At the other timepoints after baseline, a maximum of 43 cenegermin patients and 39 vehicle patients contributed data for this outcome within the FAS population.

The Schirmer test was used in ophthalmic examination to measure tear production for the diagnosis of conditions such as keratoconjunctivitis sicca and dry eye. Without previously instilling anesthetic drops, the Schirmer strip was inserted into the lower conjunctival sac at the junction of the lateral and middle thirds, avoiding touching the cornea, and the length of wetting strips in millimeters was recorded after 5 minutes. After 5 minutes had elapsed, the Schirmer test strip was removed and the length of the tear absorption on the strip was measured (millimeters/5 minutes). Cutoff values: \<5 mm - pathologic dry eye 5-10 mm - marginal dry eye \>10 and \<30 mm - normal secretion The longer, the wetted length, the healthier the status of the eye. No units other than participants were assigned.

Time frame: At Week 8 (Visit 4)

Population: Full Analysis set (FAS): The FAS population consisted of all randomized patients who received at least one dose of the investigational product. The FAS population was used for the primary analysis of the study and to present results on efficacy data. While at baseline FAS patients were n=44 for IMP and 40 for Vehicle, at week 8 patients analyzed were n=40 in the IMP group and n=38 in the Vehicle arm.

Symptoms Scores (SANDE) and/or NEI Score were punctually described in the previous outcome descriptions. For Sande score, the scale ranges from 0 to 100 for both severity and frequency, where 0 was the best condition and 100 marked the worst condition. Hence the higher the score, the worse the outcome. For NEI score, the maximum score (worst outcome) was 15, and the minimum (best outcome) was 0; hence the higher the score, the worse the outcome. Please note that mean is an adjusted mean.

Time frame: At Week 4 (Visit 3)

Population: Full Analysis set (FAS): it consisted of all randomized patients who received at least one dose of the IMP. The FAS population was used for the primary analysis and to present results on efficacy data. Patients at baseline were n=44 in the cenegermin arm and n=40 in the vehicle arm. At the other timepoints after baseline, a maximum of 43 cenegermin patients and 39 vehicle patients contributed data for this outcome within the FAS population.

Best corrected visual acuity (BCDVA) was determined by careful refraction according to the standard protocol for refraction. Chart 1 was used for testing the VA of the right eye; Chart 2 for the left eye; and Chart R for refraction only. Retroilluminated standard Early Treatment of Diabetic Retinopathy Study (ETDRS) charts were used. They had 5 Sloan letters on each line of equal difficulty, and there was a geometric progression in letter size from line to line. VAS awarded one point for every letter correctly guessed. A distance of 4 meters was required between the subject's eyes and the VA chart. When a subject cannot read at least 20 letters on the chart at 4 meters, the subject was tested at 1 meter. If 20 or more letters were read at 4 meters, the VAS for that eye was recorded as the number of letters correct at 4 meters plus 30. Otherwise, the VAS was the number of letters read correctly at 1 meter plus the number read at 4 meters. The higher the score the better the outcome.

Time frame: At Week 2 (Visit 2), Week 4 (Visit 3), Week 8 (Visit 4), Week 12 (Visit 5), and Week 16 (Visit 6)

Population: Full Analysis set (FAS): it consisted of all randomized patients who received at least one dose of the IMP. The FAS population was used for the primary analysis and to present results on efficacy data. Patients at baseline were n=44 in the cenegermin arm and n=40 in the vehicle arm. At the other timepoints after baseline, a maximum of 43 cenegermin patients and 39 vehicle patients contributed data for this outcome within the FAS population.

The NEI/Industry Workshop guidelines were used for grading the scale of corneal and conjunctival damage. The cornea was divided into five sectors (central, superior, inferior, nasal and temporal), each of which was scored on a scale of 0-3, with a maximal total corneal staining score of 15. Both nasally and temporally, the conjunctiva was divided into a superior paralimbal area, an inferior paralimbal area, and a peripheral area with a grading scale of 0-3 and with a maximal total score of 9 for the nasal and temporal conjunctiva (overall the total score ranged from 0-18). Briefly, grade 0 reflected normal/healthy situation, whereas grade 3 reflected a severe damage in the considered sector.

Time frame: At Week 2 (Visit 2)

Population: Full Analysis set (FAS): The FAS population consisted of all randomized patients who received at least one dose of the investigational product. FAS population was analysed according to intention-to-treat (ITT) principle, i.e., by treatment allocation regardless happening of intercurrent events (treatment policy strategy).~Herein the Number Participants Analyzed in Cenergermin group was 42, in the Vehicle group was 37.

The Schirmer test was used in ophthalmic examination to measure tear production for the diagnosis of conditions such as keratoconjunctivitis sicca and dry eye. Without previously instilling anesthetic drops, the Schirmer strip was inserted into the lower conjunctival sac at the junction of the lateral and middle thirds, avoiding touching the cornea, and the length of wetting strips in millimeters was recorded after 5 minutes. After 5 minutes had elapsed, the Schirmer test strip was removed and the length of the tear absorption on the strip was measured (millimeters/5 minutes). Cutoff values: \<5 mm - pathologic dry eye 5-10 mm - marginal dry eye \>10 and \<30 mm - normal secretion The longer, the wetted length, the healthier the status of the eye.

Time frame: At week 2 (Visit 2)

Population: Full Analysis set (FAS): The FAS population consisted of all randomized patients who received at least one dose of the investigational product.~While at baseline FAS patients were n=44 for IMP and 40 for Vehicle, at week 2 patients analyzed were n=43 in the IMP group and n=39 in the Vehicle arm.

Schirmer Test II (with anaesthetic) measured baseline plus reflex secretion. The Schirmer strip was inserted into the lower conjunctival sac at the junction of the lateral and middle thirds, avoiding touching the cornea, and the length of wetting strips in millimeters was recorded after 5 minutes. After 5 minutes had elapsed, the Schirmer test strip was removed and the length of the tear absorption on the strip was measured (millimeters/5 minutes). Cutoff values: \<5 mm - pathologic dry eye 5-10 mm - marginal dry eye \>10 and \<30 mm - normal secretion The longer, the wetted length, the healthier the status of the eye.

Time frame: At Week 4 (Visit 3)

Population: Full Analysis set (FAS): The FAS population consisted of all randomized patients who received at least one dose of the investigational product.~While at baseline FAS patients were n=44 for IMP and 40 for Vehicle, at week 4 patients analyzed were n=42 in the IMP group and n=38 in the Vehicle arm.

SANDE was a short questionnaire to evaluate both dry eye intensity and frequency. It uses a 100 mm horizontal line (Visual Analogue Scale - VAS), for each of the 2 questions, to assess ocular discomfort and/or dryness. Frequency of symptoms ranged from rarely (best outcome) to all of the time (worst outcome), while the severity of symptoms ranged from very mild (best outcome) to very severe (worst outcome). Patients had to place a mark on the two given lines based on the extent of their symptoms. The locations of the marks was measured in mm from left to right and recorded as frequency and severity scores, respectively. The SANDE lines for intensity and for severity ranged from 0, being the minimal amount of dry eye symptoms (best outcome) to 100, being the maximal amount of dry eye symptoms (worst outcome). The total SANDE score was calculated by multiplying the frequency score by the severity score and obtaining the square root (0-100; the lower, the better).

Time frame: At Week 2 (Visit 2) and Week 4 (Visit 3)

Population: Full Analysis set (FAS): it consisted of all randomized patients who received at least one dose of the IMP. The FAS population was used for the primary analysis and to present results on efficacy data. Patients at baseline were n=44 in the cenegermin arm and n=40 in the vehicle arm. At the other timepoints after baseline, only 43 cenegermin patients and 39 vehicle patients contributed data for this outcome within the FAS population.~Please note that mean is an adjusted mean.

Tear film break-up time (TFBUT) was the time taken to appear first dry spot on cornea after a complete blinking. TFBUT measurement was an easy and fast method used to assess the stability of tear film. It was a standard diagnostic procedure in the dry eye clinics. TFBUT was measured by determining the time to tear break-up. The TFBUT was performed after instillation of 5 μL of 2% preservative-free sodium fluorescein solution into the inferior conjunctival sac of each eye. The patient was instructed to blink several times to thoroughly mix the fluorescein with the tear film. A TFBUT greater than 15 seconds was considered normal, while a break time of less than 10 seconds was to be considered pathological.

Time frame: At Week 2 (Visit 2)

Population: Full Analysis set (FAS): The FAS population consisted of all randomized patients who received at least one dose of the investigational product. FAS population was analysed according to intention-to-treat (ITT) principle, i.e., by treatment allocation regardless happening of intercurrent events (treatment policy strategy).~Herein the Number Participants Analyzed in Cenergermin group was 43, in the Vehicle group was 39.

Symptoms Scores (SANDE) and/or NEI Score were punctually described in the previous outcome descriptions. For Sande score, the scale ranges from 0 to 100 for both severity and frequency, where 0 was the best condition and 100 marked the worst condition. Hence the higher the score, the worse the outcome. For NEI score, the maximum score (worst outcome) was 15, and the minimum (best outcome) was 0; hence the higher the score, the worse the outcome.

Time frame: At Week 2 (Visit 2)

Population: Full Analysis set (FAS): it consisted of all randomized patients who received at least one dose of the IMP. The FAS population was used for the primary analysis and to present results on efficacy data. Patients at baseline were n=44 in the cenegermin arm and n=40 in the vehicle arm. At the other timepoints after baseline, a maximum of 43 cenegermin patients and 39 vehicle patients contributed data for this outcome within the FAS population.

Treatment emergent adverse events (TEAEs) were undesirable events not present prior to medical treatment, or an already present event that worsens either in intensity or frequency following the treatment.

Time frame: From Screening day (Day -8) up to Week 24 (Visit 7 - Follow-up)

Population: The SAF set consisted of all randomized patients who received at least one dose of the investigational product. SAF set was analyzed according to the actual treatment received. The SAF population was used to present results on safety data.

Use of Preservative Free Artificial Tears by Study Period is calculated as: total number of drops of the preservative free artificial tears during the X period/ total number of days of the X period \* 100.

Time frame: Treatment period (Day 1 to Week 4), Follow-up period (Week 4 to Week 24), Overall period (Day 1 to Week 24)

Population: Full Analysis set (FAS): it consisted of all randomized patients who received at least one dose of the IMP. The FAS population was used for the primary analysis and to present results on efficacy data. Patients at baseline were n=44 in the cenegermin arm and n=40 in the vehicle arm. At the other timepoints after baseline, a maximum of 41 cenegermin patients and 38 vehicle patients contributed data for this outcome within the FAS population.